Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
In Routebase’s September 2026 study, 1,230 of 1,927 public API specification URLs returned HTTP 200, while 521 returned 404 or 410. But the larger lesson comes from a narrower comparison: among 493 self-contained specification pairs, 48.9% had a structural change, and most changed contracts included at least one change classified as breaking. A major-version number alone often did not signal those changes in this sample.
What happened to the 1,927 specification URLs?
Routebase re-fetched the origin URLs for Swagger 2.0 and OpenAPI 3.x specifications listed in the APIs.guru directory. Of the 1,927 URLs checked on September 14, 2026, 1,230 returned HTTP 200. Another 521 returned 404 or 410; 122 did not answer, while the remainder returned other outcomes, including authorization errors.
Those are URL results, not a census of whether the APIs themselves were still operating. A provider could have moved a specification file while leaving its API online. The fetch was a single attempt with a 20-second timeout and no retries, so temporary network problems could also affect the outcome.
Why the contract-change findings use only 493 pairs
The study used the APIs.guru OpenAPI directory’s list.json at commit f04b8d0bcd39c52e1cf3ad7a5fe744709832ae49, read September 14, 2026. Because the directory’s update job had last refreshed a specification in April 2023, Routebase treated its copies as a spring 2023 snapshot. The old files were the directory’s rendered and sometimes corrected copies, not necessarily byte-for-byte versions of the providers’ originals.
#1 Best Overall
- API Design Patterns
- ABIS BOOK
- Manning Publications
Of the URLs checked, 838 produced parseable documents that could be paired with a 2023 copy. The headline contract comparison narrowed that set to 493 self-contained pairs. Specifications that referenced external documents were excluded because the parser treated those references as empty objects for this comparison. Parser-rejected files were also left out of the diff and lint results; the companion repository says documents larger than 30 MB were skipped.
Routebase’s parser compared the documents for structural, consumer-visible changes, excluding description and example edits and inlining local references. That makes the result a comparison of parsed contracts, not raw file differences.
How often did the comparable contracts change?
Among the 493 pairs, 48.9% had a structural change a client could observe. Of those changed contracts, 66.8% contained at least one change Routebase classified as breaking. These percentages describe the selected, self-contained pairs—not all public APIs, or even all specifications in the directory.
Recommended Free Tools
Rank #2
Among specifications with a classified breaking change, the most common categories were:
- Endpoint removal: 45.3%.
- Response-field removal: 44.1%.
- Parameter removal: 36.6%.
- Response-type change: 36.0%.
- A newly required request-body field: 17.4%.
- Endpoint-path change: 15.5%.
A specification could contain more than one category, so these percentages overlap and should not be added together.
Did version numbers warn clients about breaking changes?
Routebase identified 161 comparable contracts with at least one breaking change. In 143 cases (88.8%), the major version did not increase; only 18 raised it. In 86 cases (53.4%), the entire version string remained unchanged. This describes the version strings in the examined sample under Routebase’s classification method. It does not establish that semantic versioning failed in general or that every provider promised a particular compatibility policy.
Did the result depend on a few large providers?
The directory was concentrated: four providers accounted for more than half of it, according to Routebase. The study checked whether those providers dominated the pattern by calculating provider-weighted and reduced-sample results.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →| Calculation | Contracts with structural change | Changed contracts with a breaking change | Breaking contracts without a major-version increase |
|---|---|---|---|
| Raw results | 48.9% | 66.8% | 88.8% |
| One vote per provider | 42.0% | 70.8% | 87.7% |
| Excluding the four largest providers | 50.3% | 66.7% | 88.8% |
The pattern persisted under both alternative calculations, though the exact structural-change percentage shifted. These checks do not make the directory representative of every public API.
What does the longer history add?
In a separate history analysis, Routebase examined 106,083 consecutive revisions of existing specification files from 2015 to 2023. It reported that 26.7% changed the contract structurally; 30.8% of those structural changes were classified as breaking. This is a different analysis from the 493-pair re-fetch and should not be combined with its percentages.
What did the 2023 specifications say about contract quality?
Routebase also assessed a separate population: 2,481 specifications in the 2023 snapshot. Under its checks, 47.6% defined a 4xx error response for at least one endpoint, 27.8% attached no security requirement to any endpoint, 20.5% defined no security scheme, and 31.8% had a version string matching the x.y.z shape it checked.
Routebase’s own neutral-core style rules flagged 98.1% of those specifications, with a median of 13 findings. Those lint findings reflect Routebase’s conventions; they are not scores from an independent standard. The 2,481-spec snapshot and the 493-pair change analysis are different populations with different denominators.
How much confidence should readers place in the findings?
This is a company-published study using Routebase’s own parser, style rules, and breaking-change definitions. The team says it manually reviewed ten randomly selected breaking-change verdicts. Its companion repository publishes pseudonymized data, tables, checksums, and an aggregation script that can recompute the reported aggregates without network access. The dataset and tables are identified as CC BY 4.0, and the script as MIT. That makes the aggregates inspectable and reproducible, but it is not an independent replication of the parser’s classifications.
The results cover public APIs represented in APIs.guru, not internal APIs or necessarily every public API. Only Swagger 2.0 and OpenAPI 3.x origin URLs were re-fetched. The URL check’s single attempt and lack of retries leave room for temporary failures, while excluding external references means the main contract analysis did not cover every paired document. Finally, “breaking” is Routebase’s operational label, not an independent standards-body ruling on a provider’s compatibility policy.
What API consumers and publishers should do
For API consumers
Do not treat a major-version number as the only warning that a contract may have changed. Review the specification diff for changes that affect your integration—such as removed endpoints or fields, changed response types, and newly required request fields—and verify important behavior against the live API.
For API publishers
Compare a proposed specification with its previous version before release, then check live behavior against the published contract. Those are practical safeguards suggested by the study’s findings; the corpus analysis did not test whether a particular product or workflow prevents incidents.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteQuick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
